Please submit both your answers and Stata code).

1.  Using insheet to load the Singapore HDB resale data resale-flat-prices- based-on-registration-date-from-jan-2017-onwards.csv and generate three new variables, price per m2 , log price per m2 , and log area

2.  Proper label the three newly created variables and existing variables re- sale price and floor area sqm.

3.  Calculate basic summary statistics for the three newly created variables and existing variables resale price and floor area sqm using the following table template. The variable columns should contains variable label rather than variable name.  Please keep 1 decimal point for all statistics.

VARIABLES    N    mean    sd    min    max

4.  Plot the histogram for price per m2 , log price per m2 , and resale price

5.  Run the following five regressions, report the regression result following the template, and interpret the coefficient on the independent variables.

reg  resale_price reg  price_m2        reg  ln_price_m2  reg  price_m2

reg  ln_price_m2

floor_area_sqm

floor_area_sqm

floor_area_sqm

ln_floor_area

ln_floor_area

Standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

6.  Use the following two commands to create a  loor variable that repre- sents the lower bound of a unit’s floor level.  Should we control floor in the regression?  Should we include a quadratic function of floor in the regression?

split  storey_range,  p("  ")

destring  storey_range1,  gen(floor)

reg  ln_price_m2  ln_floor_area  floor                           reg  ln_price_m2  ln_floor_area  floor  floor_squared

Reports the regression results in a Table following the similar format as in question 5.  What is the impact of one additional floor on the price evaluated at loor=3 from the above two regression?